The effect of elevation on species richness in tropical forests depends on the considered lifeform: results from an East African mountain forest Legrand Cirimwami, Charles Doumenge, Jean-Marie Kahindo & Christian Amani Tropical Ecology ISSN 0564-3295 Trop Ecol DOI 10.1007/s42965-019-00050-z 1 23 Your article is protected by copyright and all rights are held exclusively by International Society for Tropical Ecology. This e-offprint is for personal use only and shall not be self- archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”. 1 23 Author's personal copy Tropical Ecology International Society https://doi.org/10.1007/s42965-019-00050-z for Tropical Ecology RESEARCH ARTICLE The efect of elevation on species richness in tropical forests depends on the considered lifeform: results from an East African mountain forest Legrand Cirimwami1,2 · Charles Doumenge3 · Jean‑Marie Kahindo2 · Christian Amani4 Received: 20 March 2019 / Revised: 21 October 2019 / Accepted: 19 November 2019 © International Society for Tropical Ecology 2019 Abstract Elevation gradients in tropical forests have been studied but the analysis of patterns displayed by species richness and eleva- tion have received little attention. We examined whether the efect of elevation on species richness varies according to forest lifeforms and the main plant families in the Kahuzi-Biega National Park, within the Albertine Rift. We established 20 1-ha plots from 810 to 2760 m asl. Inside each plot, species inventories were carried out within three nested sub-plots: the tree lifeform (i.e. species with a dbh ≥ 10 cm), the shrub lifeform (dbh < 10 cm) and the herbaceous lifeform. For trees and shrubs (woody lifeforms) abundance data (i.e. number of individuals per species) were taken into account whereas the herbaceous lifeform was surveyed using presence–absence data. We plotted species counts vs elevation for each of the ten richest fami- lies per forest lifeform and resorted to Poisson regression models to assess the statistical meanings of the displayed results. Hurdle models (truncated Poisson regression) were used to account for overdispersion in the data. For woody lifeforms, we observed a monotonic decrease of species richness, while species richness appeared to be increasing with elevation for the herbaceous lifeform. Woody lifeforms displayed various vegetation patterns according to the considered families, therefore, contrasting with the general pattern observed in the herbaceous lifeform. These fndings suggest the existence of specifc eco-physiological properties pertaining to each forest lifeform and the existence of family-specifc elevation patterns of species richness. Keywords Elevation gradients · Species richness · Tree lifeform · Shrub lifeform · Herbaceous lifeform · Kahuzi-Biega national park Introduction Mountain ecosystems account for approximately one-ffth Electronic supplementary material The online version of this of the Earth’s land surface (Tse-ring et al. 2010; Price et al. article (https ://doi.org/10.1007/s4296 5-019-00050 -z) contains 2011) and tropical mountain forests represent about 11% supplementary material, which is available to authorized users. of tropical forests over the world (Doumenge et al. 1995). * Legrand Cirimwami Species richness is particularly high in these ecosystems [email protected] and they are often biodiversity hotspots (Bussmann 2001; 1 Plumptre et al. 2003; Tse-ring et al. 2010; Price et al. 2011; Faculté des Sciences, Université du Cinquantenaire, Lwiro, Admassu et al. 2016). In these mountain ecosystems, eleva- Sud-Kivu, Democratic Republic of the Congo 2 tion strongly infuences the distribution of their biodiversity Faculté des Sciences, Université de Kisangani, PO Box. 2012, (Doumenge 1998; Eilu et al. 2004; Hemp 2009; Pickering Kisangani, Tshopo, Democratic Republic of the Congo 3 and Green 2009; Chain-Guadarrama et al. 2012). Indeed, UR Forêts et Sociétés, CIRAD, Campus International de elevation regulates several abiotic factors that control the Baillargue TA-C-105/D, F-34398 Cedex 5 Montpellier, France observed vegetation patterns and the ecology of moun- tain forests. Climatic variables such as temperature or 4 Département de Biologie, Faculté des Sciences et Sciences Appliquées, Université Ofcielle de Bukavu, PO Box. 570, atmospheric pressure decrease with increasing elevation Bukavu, Democratic Republic of the Congo while radiation under cloudless sky and fraction of UV-B Vol.:(0123456789)1 3 Author's personal copy Tropical Ecology radiation at any given total solar radiation increase (Körner (Poulsen and Pendry 1995; Salinas and Casas 2007; Nanda 2007; Chain-Guadarrama et al. 2012). Other factors such as et al. 2018) as well as along large elevational gradients on light availability (in terms of difuse transmittance), cloudi- Mount Kilimandjaro (Hemp 2006). This tendency is con- ness and humidity (Lebrun and Gilbert 1954; Körner 2007; frmed by a transcontinental (America, Africa and South Unger et al. 2013), soil depth (Doležal and Šrŭtek 2002) and East Asia) study that showed that species richness of under- soil parameters (e.g. organic matter, total nitrogen, water story herbs in tropical forests is positively correlated with content, available nitrogen and C/N ratio) increase, whereas increasing elevation (Cicuzza et al. 2013). The elevation pH, N mineralization and nitrifcation tend to decrease or patterns of diversity for herbaceous lifeforms are explained present an intermediate maximum with increasing eleva- by light availability as most of them are shade intolerant tion, depending on the climatic context and elevational range (Pierlot 1966; Grytnes 2000; Jiang et al. 2016). (Zhang et al. 2012). Furthermore, the orographic character- The relationship between elevation and species richness istics diferently afect precipitation in mountain ecosystems, may also vary according to the considered family (Kessler resulting into local variations in species composition. 2000; Lomolino 2001; Salas-Morales and Meave 2012). In Facing all these factors driven by elevation, species from many studies in which diferent families’ species richness diferent taxa, families or lifeforms respond specifcally were plotted according to elevation, a wide variety of pat- according to their eco-physiological properties and sensi- terns was observed (Kessler 2000, 2002; Delnatte 2010; tivity (Grime 1977, 1979; Kessler 2000; Pausas and Austin Salas-Morales and Meave 2012). The main explanations of 2001; Chawla et al. 2008). For example, several studies have the existence of this wide variety of patterns are eco-physio- found that, among functional traits, solar radiation tolerance logical properties of families and their functional traits. Fur- is one of the explanatory factors shaping diversity patterns thermore, the elevational patterns of species richness within in forest understory layers (Grytnes 2000, Willinghöfe et al. diferent families per lifeform (herbs to trees) may clearly 2012, Nanda et al. 2018). difer from that of the whole plant community (Chawla et al. Some studies found that the relationship between tree spe- 2008; Salas-Morales and Meave 2012). cies richness (for species with at least 10 cm of diameter at Few studies compared the elevational variation in diver- breast height) and elevation is unimodal, displaying a peak sity and composition of various plant lifeforms, from herbs around 1000 m asl (Pierlot 1966; Doumenge 1998; Bhattarai to trees in mountains of tropical Africa (Woldu et al. 1989; and Vetaas 2006; Lee et al. 2013), while others reported that Hemp 2006; Fischer et al. 2011). However, this information tree species richness declines monotonically with increas- is nevertheless key for conservation as well as REDD + strat- ing elevation (Stevens 1992; Zhao et al. 2004; Kraft et al. egies in a context of climate change (Foster 2001; Malhi 2011; Zhu et al. 2015). Most of these studies were performed et al. 2010; Malhi and Marthews 2013). The importance in the temperate zone and few data are available on eleva- of such studies for REDD + strategies is due to the fact that tional gradients in tropical Africa (Pierlot 1966; Imani et al. several components of REDD + (e.g. the measure of Above 2016; Cirimwami et al. 2017). In the East African Moun- Ground Biomass) are infuenced by the species richness of tains, when considering wide elevational gradients, patterns studied sites (Gonmadje et al. 2017; Imani et al. 2017). of tree species richness tend to present a maximum in the In this study, we focused on how species richness of three lower elevations of montane forests (Pierlot 1966; Doume- lifeforms (tree, shrub and herbaceous) varies with eleva- nge 1998; Schmitt et al. 2010). tion—810 to 2760 m asl—in the Kahuzi-Biega National Although the relationship between species richness and Park (KBNP), East of the Democratic Republic of the Congo elevation has been shown to be a function of the vegetation (DRC). We hypothesized that (1) the relationship between lifeform studied